Senior AI Engineer (Driving VLM/VLA)

Posted 14 Days Ago
Be an Early Applicant
Hiring Remotely in Pangyo-ri, Gangweon Do, KOR
Remote or Hybrid
Senior level
Artificial Intelligence • Software • Transportation
The answer to mobility and everything
The Role
Design and develop vision-language-action driving foundation models using multimodal sequential data. Train and evaluate large-scale models for driving scene understanding, temporal reasoning, agent interaction, and risk/event understanding. Build VLM-based applications (captioning, VQA, auto-labeling), construct multimodal datasets, design training recipes and benchmarks, run ablations, and collaborate with Simulation, Data, and ML Platform teams to integrate models into autonomous driving systems.
Summary Generated by Built In
We are looking for the best

42dot의 Senior AI Engineer (Driving VLM/VLA)는 수백만 건의 실제 자율주행 데이터를 활용해 차세대 Driving Foundation Model을 개발합니다. Vision-Language-Model(VLM), Vision-Language-Action(VLA), 멀티모달 AI 기술을 통해 차량이 복잡한 도로 환경을 이해하고 판단하며 행동할 수 있도록 만드는 핵심 역할을 수행합니다. 대규모 GPU 인프라와 독자적인 자율주행 데이터셋을 기반으로 driving scene understanding, multimodal reasoning, VLM/VLA를 통해 자율주행 시스템의 인지, 판단, 계획 성능을 고도화합니다.

Responsibilities

  • 자율주행 데이터를 활용한 Vision-Language-Action 기반 driving foundation model 설계 및 개발

  • Camera/video, map, trajectory, action, language 등 multimodal sequential data 기반 모델 학습 및 평가

  • Driving scene understanding, temporal reasoning, agent interaction modeling, risk/event understanding 모델 개발

  • Scene captioning, visual question answering, auto-labeling, data mining, retrieval 등 VLM 기반 응용 기능 개발

  • 대규모 multimodal dataset 구축, 학습 recipe, evaluation benchmark, ablation 실험 설계

  • Simulation, Data, ML Platform 조직과 협업하여 모델을 자율주행 시스템에 통합

Qualifications

  • Computer Vision, Machine Learning, Robotics, Autonomous Driving, Multimodal AI 관련 5년 이상의 연구/개발 경험 또는 이에 준하는 역량 또는 Foundation Model, Multimodal Learning, Generative AI 분야에서의 뛰어난 연구 성과(proven track record)를 보유하신 분

  • PyTorch 기반 deep learning model 개발 및 학습 경험

  • Vision model, video model, VLM, VLA, world model, trajectory prediction, imitation learning 중 하나 이상에 대한 깊은 이해와 구현 경험

  • 이미지/비디오/sensor/trajectory/language/action 등 multimodal 또는 sequential data 처리 경험

  • Transformer, diffusion, autoregressive model, representation learning 중 하나 이상에 대한 이해

  • 논문 기반 아이디어를 구현하고 대규모 데이터에서 학습, 평가, 개선한 경험

  • 문제를 독립적으로 정의하고, 실험 설계부터 모델 개선까지 주도적으로 수행할 수 있는 역량

Preferred Qualifications

  • 자율주행 또는 robotics foundation model 개발 경험

  • VLM/LLM fine-tuning, instruction tuning, multimodal alignment 경험

  • VLA, behavior cloning, imitation learning, trajectory planning, policy learning 경험

  • Action-conditioned video generation, world model, 4D scene modeling 경험

  • BEV, occupancy, map, trajectory, agent interaction 등 driving-specific representation 경험

  • Distributed training, large-scale multimodal data pipeline, data mining pipeline 경험

  • Closed-loop simulation 또는 scenario-based evaluation과 모델을 연결해본 경험

Interview Process

  • 서류전형 - 코딩테스트 - 화상면접 (1시간 내외) - 대면 혹은 화상면접 (3시간 내외) - 최종합격

  • 전형절차는 직무별로 다르게 운영될 수 있으며, 일정 및 상황에 따라 변동될 수 있습니다.

  • 전형일정 및 결과는 지원서에 등록하신 이메일로 개별 안내드립니다.

Additional Information

  • 이력서 제출 시 주민등록번호, 가족관계, 혼인 여부, 연봉, 사진, 신체조건, 출신 지역 등 채용절차법상 요구 금지된 정보는 제외 부탁드립니다.

  • 모든 제출 파일은 30MB 이하의 PDF 양식으로 업로드를 부탁드립니다. (이력서 업로드 중 문제가 발생한다면 지원하시고자 하는 포지션의 URL과 함께 이력서를 [email protected]으로 전송 부탁드립니다.)

  • 인터뷰 프로세스 종료 후 지원자의 동의하에 평판조회가 진행될 수 있습니다.

  • 국가보훈대상자 및 취업보호 대상자는 관계법령에 따라 우대합니다.

  • 장애인 고용 촉진 및 직업재활법에 따라 장애인 등록증 소지자를 우대합니다.

  • 42dot은 의뢰하지 않은 서치펌의 이력서를 받지 않으며, 요청하지 않은 이력서에 대해 수수료를 지불하지 않습니다.

  • 3개월의 수습기간이 적용될 수 있습니다.

※ 지원 전 아래 내용을 꼭 확인해 주세요.

  • 42dot이 일하는 방식, 42dot Way 보러가기 →

  • 42dot만의 업무몰입 프로그램, Employee Engagement Program 보러가기 →

Skills Required

  • 5+ years of R&D experience in Computer Vision, Machine Learning, Robotics, Autonomous Driving, or Multimodal AI (or equivalent proven track record)
  • PyTorch-based deep learning model development and training experience
  • Deep understanding and implementation experience in at least one: vision model, video model, VLM, VLA, world model, trajectory prediction, imitation learning
  • Experience processing multimodal or sequential data (image/video/sensor/trajectory/language/action)
  • Understanding of Transformer, diffusion, autoregressive models, or representation learning
  • Experience implementing research ideas and training, evaluating, and improving models on large-scale data
  • Ability to define problems independently and lead experiment design through model improvements
  • Autonomous driving or robotics foundation model development experience
  • VLM/LLM fine-tuning, instruction tuning, multimodal alignment experience
  • VLA, behavior cloning, imitation learning, trajectory planning, or policy learning experience
  • Action-conditioned video generation, world model, or 4D scene modeling experience
  • Driving-specific representation experience (BEV, occupancy, map, trajectory, agent interaction)
  • Distributed training, large-scale multimodal data pipeline, or data mining pipeline experience
  • Closed-loop simulation or scenario-based evaluation integrated with models
Am I A Good Fit?
beta
Get Personalized Job Insights.
Our AI-powered fit analysis compares your resume with a job listing so you know if your skills & experience align.

The Company
HQ: Seoul
739 Employees
Year Founded: 2019

What We Do

We envision a world where everything is connected and moves autonomously through a self-managing urban transportation operating system. We are spearheading the transition to SDV (software-defined vehicle) with software and AI. We are developing diverse SDV technologies that continuously provide vehicle updates and services based on data for user-centric and safe mobility. If you want to create a new future of mobility through software, AI, and automotive vehicles, check out our job openings. Come ride with us!

Why Work With Us

Similar Jobs

SailPoint Logo SailPoint

Account Executive

Artificial Intelligence • Cloud • Sales • Security • Software • Cybersecurity • Data Privacy
Remote or Hybrid
South Korea
2461 Employees

CrowdStrike Logo CrowdStrike

Regional Sales Manager

Cloud • Computer Vision • Information Technology • Sales • Security • Cybersecurity
Remote or Hybrid
South Korea
10000 Employees

Tapestry - Coach and Kate Spade Logo Tapestry - Coach and Kate Spade

Sales Associate

eCommerce • Fashion • Retail • Sales • Wearables • Design
Remote or Hybrid
Si-dong, Gangweon Do, KOR
16000 Employees

Zscaler Logo Zscaler

Senior Manager, Sales Engineering

Cloud • Information Technology • Security • Software • Cybersecurity
Easy Apply
Remote or Hybrid
South Korea
8697 Employees

Similar Companies Hiring

Hanover Park Thumbnail
Artificial Intelligence • Fintech • Software • Financial Services
New York, New York
42 Employees
Kepler  Thumbnail
Fintech • Software
New York, New York
6 Employees
Onshore Thumbnail
Artificial Intelligence • Fintech • Software • Financial Services
New York, New York
60 Employees

Sign up now Access later

Create Free Account

Please log in or sign up to report this job.

Create Free Account